The 2006 report you refer to is the last in a series of three. The pattern has been tracked across two decades, with surveys conducted by two separate polling organizations.
- 1988 - performed by Gallup, sampling 18-24 year olds (New York Times article)
- 2002 - performed by Roper, sampling 18-24, and 25-35 year olds in the US, and compared against the 1988 results (report)
- 2006 - performed by Roper, sampling 18-24 year olds in the US (report)
1988 Highlights (from the NYT article)
- no more than half those tested could answer correctly when asked to identify the country in which the Sandinistas and the contras are fighting
- 52 percent were aware that the Soviet Union's war with rebel forces was in Afghanistan
- 55 percent could identify South Africa as the country where apartheid is official government policy
- only 25 percent could name four countries that officially acknowledge having nuclear weapons
- more than a third could not pick out the westernmost city on a map
- just 23% of young adults surveyed could name
four countries that officially acknowledge having nuclear weapons
- young adults in other countries in the study were better able to locate other
European countries than American young adults are to locate U.S. states
- close to onethird
(30%) said that the U.S. has 1 to 2 billion people, or roughly one-third of the world s
population. Only one-fourth of American young adults (25%) correctly identified the U.S.
population as falling within the range of 150 to 350 million. Respondents in all other
countries were better able to identify the U.S. population than are young Americans.
- just 17% could find Afghanistan on a world map.
- only 14% of young Americans could locate Israel on a Middle East/Asia map
- young Americans did seem more geographically literate on current events that occur near or in the U.S. or that affect U.S. life
- When asked to identify the western-most city on a hypothetical map, seven in ten young Americans (70%) answered correctly — up 12 points from 1988.
- When asked to identify 12 countries on a map of Europe, the average young American
could locate only about 3, compared to their counterparts in the other countries
surveyed, who could locate an average of 5.
- Just 37% could find England
- I'm not going to repeat all of the results, but some highlights are listed here (which you link to in the question).
To address your specific doubt about the representativeness of the sample, see p. 46-47 of the 2006 report. The sample was not intended to be representative of any particular state (and some states might have been poorly represented). The sample was intended to be representative of the 18-24 year old population in the continental United States.
A multistage, stratified area probability sample applied down to the Secondary Sampling Unit (SSU) stage (i.e. street/block group or similar classification) was used for this research. For the selection of households, interviewers followed randomly selected, pre-determined interviewing routes with skip intervals. At the household level, quotas for sex and age were applied for respondent selection.
Expectation of more than one paper
You expect more than one paper about American geographic literacy. There are many:
- Carano, Kenneth T., and Michael J. Berson. "Breaking stereotypes: Constructing geographic literacy and cultural awareness through technology." The Social Studies 98, no. 2 (2007): 65-69.
- Lukinbeal, Chris, and Jim Craine. "Geographic media literacy: an introduction." GeoJournal 74, no. 3 (2009): 175-182.
- Gaudelli, William, and Elizabeth Heilman. "Reconceptualizing geography as democratic global citizenship education." The Teachers College Record 111, no. 11 (2009): 2647-2677.
- Turner, Sally, and Joseph Leydon. "Improving Geographic Literacy among First-Year Undergraduate Students: Testing the Effectiveness of Online Quizzes." Journal of Geography 111, no. 2 (2012): 54-66.
There are multiple papers on the topic. They take their motivation from this series of three surveys.
A more recent survey
The American Geographic Society's Geographic Knowledge and Values Survey
This was an online survey, with volunteers soliciting responses from US residents. They had 4021 valid responses. This was not a random sample, and is not representative of anything other than the population of people who tend to answer surveys like this. Respondents had a high level of education.
- Of the 3,972 respondents who answered regarding the axis on which longitude changes, 56% of the respondents answered incorrectly [...] This result is
little better than what might be expected from random guesses.
- The majority of the respondents
said they did not know if a conformal or Mercator projection was appropriate for comparing the
land areas of Russia and Kenya. For those who did choose an answer, 32.3% incorrectly said
that a conformal projection was appropriate, while 34.8% correctly said that a Mercator
projection would not be appropriate
- The majority of respondents, 55.4%, correctly chose a map scale of 1:50,000,000 for a
map of the entire earth at page size, compared to 14.7% who chose 1:5,000 and 30% who said
they did not know. [...] this result is little better than what might be
expected from random guesses
- 23.3% correctly identified Florida as the flattest state
- Most respondents (67.8%) correctly chose Africa South of the Sahara as the
poorest world region
In my opinion, this survey does not overturn the results found in 1988, 2002, and 2006. It asked a very limited set of geographic literacy questions, and its sample was not representative of young Americans.